TY - GEN
T1 - A BDS/GPS Integrated Positioning Algorithm Based On Kalman Filter and Deep Neural Networks
AU - Jiang, Xia
AU - Li, Qingli
AU - Wei, Decheng
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - As fueled by the advancement of machine learning, the application of neural networks, a branch of machine learning, has been increasingly extensive. Meantime, with the advent of BDS, any position worldwide can be fixed with BDS/GPS. Though BDS/GPS exhibits high accuracy, a more accurate navigation by integrating BDS/GPS positioning is required. Thus, a dual model of integrating BDS/GPS navigation data is introduced in this study. To exploit BDS/GPS and dual model, a novel integrated positioning algorithm combining the Kalman filter and deep neural networks is proposed to simulate the navigation and positioning. The input of our paradigm originates from BDS/GPS and dual model. Stability of raw positioning data can be enhanced by Kalman filtering. Subsequently, a deep neural network is built to integrate the simulations. As revealed from experimental results, stabilization of the input data can enhance reliability of the deep neural network; besides, the proposed data integration algorithm can enhance the accuracy and stability of the navigation and positioning.
AB - As fueled by the advancement of machine learning, the application of neural networks, a branch of machine learning, has been increasingly extensive. Meantime, with the advent of BDS, any position worldwide can be fixed with BDS/GPS. Though BDS/GPS exhibits high accuracy, a more accurate navigation by integrating BDS/GPS positioning is required. Thus, a dual model of integrating BDS/GPS navigation data is introduced in this study. To exploit BDS/GPS and dual model, a novel integrated positioning algorithm combining the Kalman filter and deep neural networks is proposed to simulate the navigation and positioning. The input of our paradigm originates from BDS/GPS and dual model. Stability of raw positioning data can be enhanced by Kalman filtering. Subsequently, a deep neural network is built to integrate the simulations. As revealed from experimental results, stabilization of the input data can enhance reliability of the deep neural network; besides, the proposed data integration algorithm can enhance the accuracy and stability of the navigation and positioning.
KW - BDS
KW - GPS
KW - Kalman filter
KW - deep neural networks
KW - dual model
UR - https://www.scopus.com/pages/publications/85183317460
U2 - 10.1109/CISP-BMEI60920.2023.10373356
DO - 10.1109/CISP-BMEI60920.2023.10373356
M3 - 会议稿件
AN - SCOPUS:85183317460
T3 - Proceedings - 2023 16th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2023
BT - Proceedings - 2023 16th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2023
A2 - Zhao, XiaoMing
A2 - Li, Qingli
A2 - Wang, Lipo
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 16th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2023
Y2 - 28 October 2023 through 30 October 2023
ER -